Note: Bibliographic details of the publications referred to by author in this specification are collected at the end of the description.
The acquisition of the malignant phenotype involves a number of initiation and progression factors linked together in a multi-step process (Meyer and Hart 1998; Haber and Fearon, 1998). Malignant tumors are potentially lethal because of the ability of cells within the tumor to invade and spread (metastasize) throughout the body (Meyer and Hart 1998). The process of metastasis can be summarized in the following steps; escape of tumor cells from a primary tumor mass and invasion into surrounding histologically normal tissue, intravasation and extravasation (entry into and exit from the vasculature or lymphatic system), and growth and survival (via angiogenesis) of tumor cells at a secondary site. One of the major failures in the treatment of cancer is poor detection and eradication of metastases before vital organ functions are compromised, resulting in minimal long-term survival benefit (Allen, 1999). Thus, effective cancer treatment entails primary tumor resection followed by removal of all metastases arising from the primary. The latter would require (1) identification of a marker specific to metastatic cells, and (2) development of a ligand that can be used to specifically target and kill the cells expressing the marker.
Proteolytic enzymes such as urokinase plasminogen activator (herein referred to as “uPA”) play a role in tumor angiogenesis and metastatic cell migration; both of which are processes that require tissue barriers to be breached (reviewed in Andreasen et al., 1997). Under normal physiological conditions, most cells express very little or no uPA (Pollanen et al., 1991). Urokinase plasminogen activator converts zymogen plasminogen into the highly active protease plasmin, which has broad specificity towards integral extracellular matrix (ECM) molecules (eg: type IV collagen, vitronectin, proteoglycan, fibronectin and laminin) (Pollanen et al., 1991). Plasmin also contributes to ECM remodeling by activating zymogen metalloproteases (MMPs) which more thoroughly degrade the collagen structural components (Pollanen et al., 1991). While plasminogen can also be converted to plasmin by tissue plasminogen activator (tPA), tPA is primarily responsible for fibronolysis (Lijnen and Collen, 1982; Pollanen et al., 1991). In contrast uPA is primarily involved in pericellular proteolysis as it binds to its specific cell-surface receptor uPAR (Pollanen et al., 1991). The activities of uPA and plasmin are physiologically inhibited by the serpins plasminogen activator inhibitors type 1 and 2 (PAI-1 and PAI-2) and alpha 2-antiplasmin, respectively (Pollanen et al., 1991). The uPA system and MMPs, such as MMP-9, have been shown to act cooperatively in allowing tumor cells to breach the vascular wall (Kim et al., 1999).
Prior art “magic bullet” style treatments for neoplastic conditions have been extensively investigated but, to date, have met with little or no success.
Specifically, most previous attempts at such targeted treatment have relied on the use of antibodies directed to various cell surface molecules expressed by cancer cells. However, such approaches have suffered from many drawbacks including:                (i) the surface molecules to which the antibodies are directed have not been uniquely expressed by the cancer cells. Accordingly, such treatments have also been toxic to a significant number of normal cells.        (ii) the antibodies which have been utilized in such treatments have been of mouse origin or have been “humanized mouse antibodies. The use of such antibodies has led to immunological complications associated with the HAMA response. That is, repeated dosing of humans with murine antibodies, such as humanized antibodies, has led to the development of anti-murine antibodies thereby causing both rapid clearance of the murine antibodies and the production of immune complexes which cause the HAMA response.        (iii) the antibody-cell surface molecule complexes are often internalized. Accordingly, the toxin which is coupled to the subject antibody is less effective.        (iv) the strength of binding of most antibodies to a target cell surface molecule is low with a dissociation constant of approximately 10−6M being common.        (v) in coupling a toxin to the antibody the point of linkage cannot usually be predicted. Depending on the structure of a given antibody, coupling may occur at the antigen binding site region of the antibody thereby rendering the antibody useless.        
Further, prior art radiocolloid therapy is not suitable for adjunctive therapy as it is not selective of cancer cells. To the extent that beta and gamma emitting radionuclides have been coupled to specific monoclonal antibodies, problems have been experienced with most of the dose leaving the cancer cell. Therefore, therapeutic doses cannot be achieved without inducing severe complications.
Accordingly, there is a need to develop a more effective method of targeting neoplastic cells for treatment, which method provides both improved selectivity in terms of its targeting function and improved delivery of a toxic signal. In terms of the delivery of a toxic signal, there is a need to develop a method which provides both a maximal dose of the subject toxin to the target cell but with minimal impact upon proximally located non-target cells.